Large Language Models (LLMs) such as ChatGPT have demonstrated remarkable proficiency on various natural language processing (NLP) tasks such as generating plans, writing source code, and interpreting results. However, to solve complex problems in domains such as physics, math, and engineering, additional intelligence is required. In this seminar, I will explore new developments and ideas for combining LLMs with advanced tools and libraries, leveraging their strengths to overcome challenges and tackle complex tasks.